Carcinogenicity Classification: Para Isomer (CAS 2834-78-8) Is Experimentally Non-Carcinogenic vs. Carcinogenic Ortho Isomer (CAS 97-56-3)
In a QSAR study by Song et al. (2016) that evaluated the carcinogenic potency of 93 aromatic amines using gene expression programming (GEP) and multilayer perceptron (MLP) models, 4-(p-tolylazo)-o-toluidine (entry 89) was experimentally classified as non-carcinogenic (carcinogenicity (exp) = 0), whereas its ortho-tolyl positional isomer 4-(o-tolylazo)-o-toluidine (entry 86) was experimentally confirmed as carcinogenic (carcinogenicity (exp) = 1) [1]. Both GEP and MLP computational models also predicted the para isomer as non-carcinogenic (both scored 0), while the ortho isomer was predicted carcinogenic by GEP (score 1) [1]. This experimental and computational concordance establishes a clear, quantifiable differentiation between the two positional isomers.
| Evidence Dimension | Experimental Carcinogenicity Classification |
|---|---|
| Target Compound Data | Carcinogenicity (exp) = 0 (non-carcinogenic); GEP predicted = 0; MLP predicted = 0 |
| Comparator Or Baseline | 4-(o-Tolylazo)-o-toluidine (CAS 97-56-3): Carcinogenicity (exp) = 1 (carcinogenic); GEP predicted = 1; MLP predicted = 0 |
| Quantified Difference | Binary classification difference: non-carcinogenic vs. carcinogenic (exp: 0 vs. 1) |
| Conditions | QSAR study dataset of 93 aromatic amines; experimental carcinogenicity sourced from CPDB and published literature; GEP model accuracy 0.92 (training) / 0.82 (test); MLP accuracy 0.84 (training) / 0.74 (test) |
Why This Matters
This carcinogenicity differential determines regulatory permissibility under EU Regulation 552/2009/EC: the ortho isomer is banned from textile and leather articles, while the para isomer's non-carcinogenic profile enables its use in applications where regulatory restrictions on carcinogenic aromatic amines apply.
- [1] Song, F.; Zhang, A.; Liang, H.; Cui, L.; Li, W.; Si, H.; Duan, Y.; Zhai, H. QSAR Study for Carcinogenic Potency of Aromatic Amines Based on GEP and MLPs. Int. J. Environ. Res. Public Health 2016, 13, 1141. Table 1, entries 86 and 89. View Source
